FAI AI Infra Expert
AI infrastructure expert — GPU compute sizing (A100/H100), VRAM estimation, model serving (vLLM/TensorRT-LLM/Triton), AKS node pool design, PTU vs PAYG cost modeling, and quantization strategies.
Overview
| Property | Value |
|---|---|
| Type | Agent |
| File | agents/fai-ai-infra-expert.agent.md |
| Tools | codebase, terminal, azure |
| Model | gpt-4o, gpt-4o-mini |
| WAF Alignment | performance-efficiency, cost-optimization, reliability |
| Compatible Plays | 02-ai-landing-zone, 12-model-serving-aks |
| Lines | 107 |
Usage
In VS Code (GitHub Copilot)
@fai-ai-infra-expert How can you help me?
In fai-manifest.json
{
"primitives": {
"agents": ["../../agents/fai-ai-infra-expert.agent.md"]
}
}
Source
- GitHub:
agents/fai-ai-infra-expert.agent.md - Edit: Edit on GitHub →
Auto-generated from the FrootAI primitive catalog. Last updated: 2026-04-20.